Religious News Service would like you and I to believe that.
One basic, Mack Truck-sized loophole, though?
"Scientist" is nowhere defined, other than, possibly, but probably not, by membership in the American Academy for the Advancement of Science, whose findings are nowhere linked to in the piece. The AAAS did the research, but it's not clear who all was researched.
That only 20 percent of the general public attends worship weekly? Yes, using time logs rather than self-reporting, that's been confirmed. Question: were scientists surveyed the same way? I don't know, because I don't see a link. So, not only do I not know who was researched, I don't know how the research was done, and if we're comparing apples to apples or not.
Second, what counts as religious services? Unitarianism and Reform Judaism are at least somewhat non-metaphysical. Indeed, Judaism is higher among scientists than the general public.
So, claiming 18 percent of scientists attend services weekly? It's almost contentless. It's definitely apples to oranges, per that second link.
Plus, exactly the opposite of the general public (at least the general public in the past), scientific religiosity decreases with age. Religiosity also decreases by specific science field, with the "harder" sciences being less religious. Again, while RNS linked to that second piece, it didn't note that it undercuts almost all of what it is trying to claim.
Third, we have further evidence to the contrary. Some 92 percent of members of the National Academy of Sciences, for which a Ph.D. is a de facto membership requirement, are atheists or agnostics. Of the remaining 8 percent, even if 100 percent attend a religious service weekly (highly doubtful), that's just 8 percent. And, that was back in 1998.
And, the general public, when thinking about "scientists," is likely to be thinking about "hard" scientists doing Ph.D. level research, thus, much closer to the NAS than to whatever undefined standard Religious News Service has.
Given that, outside of behavioral psychology, large tracts of psychology aren't that scientific, even at the Ph.D. level, and that's somewhat true of sociology, I am going to lean toward Gould's 1998 link above.
In one area that runs more religious than "hard" scientists, mathematics, I don't consider mathematics to be scientific. Rather, it is its own field. It can be put to use for cosmology and astrophysics, yes. It can also, under the old "lies, damned lies and statistics," of Twain, Disraeli or whomever, be put to use for unscientific psychology and sociology, or even more unscientific economics.
And, the fact that Chris Stedman of Faitheist fame retweets the RNS story without comment reinforces that, while he may be an evangelist for his "liberal Protestant" version of atheism, a skeptic he is not. Even if using Gould's NAS "greater scientists" angle is too narrow, whatever unexplained definition AAAS used is surely too wide.
Of course, this is all part of a broad tussle, that, speaking of Gould, goes back to his idea of "non-overlapping magisteria."
In biology, for example, evolution by natural selection is not the same as abiogenesis, though Darwin's "warm little pond" comment could be seen as including abiogenesis. And, in cosmology, no a Big Bang doesn't exclude a deity, especially if one notes that current physics understanding can't get on "the other side" of Planck time.
On issues of science education, I have no problem, then, in atheists working with folks like the National Center for Science Education, just as I personally have no problem firing off the occasional First Amendment email drafted by Americans United for Separation of Church and State.
I'm an "accommodationist" in many Gnu Atheist eyes. So sue me, P.Z. Myers.
That said, I think Gould's idea, overall, is wrong. It's little more than the old "god of the gaps" in new drag.
And, in a Twitter exchange, RNS author Cathy Lynn Grossmann doubled down on this being a PR piece, claiming the NAS definition of scientist is more subjective than the no-definition "whatever" of her piece.